Class/Object

com.intel.analytics.zoo.pipeline.api.keras.layers

GRU

Related Docs: object GRU | package layers

Permalink

class GRU[T] extends Recurrent[T] with Net

Gated Recurrent Unit architecture. The input of this layer should be 3D, i.e. (batch, time steps, input dim).

When you use this layer as the first layer of a model, you need to provide the argument inputShape (a Single Shape, does not include the batch dimension).

T

Numeric type of parameter(e.g. weight, bias). Only support float/double now.

Linear Supertypes
Net, Recurrent[T], bigdl.nn.keras.Recurrent[T], KerasLayer[Tensor[T], Tensor[T], T], Container[Tensor[T], Tensor[T], T], AbstractModule[Tensor[T], Tensor[T], T], InferShape, Serializable, Serializable, AnyRef, Any
Ordering
  1. Alphabetic
  2. By Inheritance
Inherited
  1. GRU
  2. Net
  3. Recurrent
  4. Recurrent
  5. KerasLayer
  6. Container
  7. AbstractModule
  8. InferShape
  9. Serializable
  10. Serializable
  11. AnyRef
  12. Any
  1. Hide All
  2. Show All
Visibility
  1. Public
  2. All

Instance Constructors

  1. new GRU(outputDimension: Int, activation: KerasLayer[Tensor[T], Tensor[T], T] = null, innerActivation: KerasLayer[Tensor[T], Tensor[T], T] = null, returnSeq: Boolean = false, goBackward: Boolean = false, wRegularizer: Regularizer[T] = null, uRegularizer: Regularizer[T] = null, bRegularizer: Regularizer[T] = null, mInputShape: Shape = null)(implicit arg0: ClassTag[T], ev: TensorNumeric[T])

    Permalink

    outputDimension

    Hidden unit size. Dimension of internal projections and final output.

    activation

    Activation function to use. You can also pass in corresponding string representations such as 'relu' or 'sigmoid', etc. for simple activations in the factory method. Default is 'tanh'.

    innerActivation

    Activation function for inner cells. You can also pass in corresponding string representations such as 'relu' or 'sigmoid', etc. for simple activations in the factory method. Default is 'hard_sigmoid'.

    returnSeq

    Whether to return the full sequence or only return the last output in the output sequence. Default is false.

    goBackward

    Whether the input sequence will be processed backwards. Default is false.

    wRegularizer

    An instance of Regularizer, (eg. L1 or L2 regularization), applied to the input weights matrices. Default is null.

    uRegularizer

    An instance of Regularizer, applied the recurrent weights matrices. Default is null.

    bRegularizer

    An instance of Regularizer, applied to the bias. Default is null.

    mInputShape

    A Single Shape, does not include the batch dimension.

Value Members

  1. final def !=(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  2. final def ##(): Int

    Permalink
    Definition Classes
    AnyRef → Any
  3. final def ==(arg0: Any): Boolean

    Permalink
    Definition Classes
    AnyRef → Any
  4. def accGradParameters(input: Tensor[T], gradOutput: Tensor[T]): Unit

    Permalink
    Definition Classes
    KerasLayer → AbstractModule
  5. val activation: KerasLayer[Tensor[T], Tensor[T], T]

    Permalink

    Activation function to use.

    Activation function to use. You can also pass in corresponding string representations such as 'relu' or 'sigmoid', etc. for simple activations in the factory method. Default is 'tanh'.

  6. def apply(name: String): Option[AbstractModule[Activity, Activity, T]]

    Permalink
    Definition Classes
    Container → AbstractModule
  7. final def asInstanceOf[T0]: T0

    Permalink
    Definition Classes
    Any
  8. var bRegularizer: Regularizer[T]

    Permalink

    An instance of Regularizer, applied to the bias.

    An instance of Regularizer, applied to the bias. Default is null.

  9. def backward(input: Tensor[T], gradOutput: Tensor[T]): Tensor[T]

    Permalink
    Definition Classes
    AbstractModule
  10. var backwardTime: Long

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  11. def build(calcInputShape: Shape): Shape

    Permalink
    Definition Classes
    KerasLayer → InferShape
  12. def buildCell(input: Array[Int]): Cell[T]

    Permalink
    Definition Classes
    GRU → Recurrent
  13. def canEqual(other: Any): Boolean

    Permalink
    Definition Classes
    Container → AbstractModule
  14. final def checkEngineType(): GRU.this.type

    Permalink
    Definition Classes
    Container → AbstractModule
  15. def clearState(): GRU.this.type

    Permalink
    Definition Classes
    Container → AbstractModule
  16. final def clone(deepCopy: Boolean): AbstractModule[Tensor[T], Tensor[T], T]

    Permalink
    Definition Classes
    AbstractModule
  17. def clone(): AnyRef

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  18. final def cloneModule(): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  19. def computeOutputShape(inputShape: Shape): Shape

    Permalink
    Definition Classes
    Recurrent → KerasLayer → InferShape
  20. def doBuild(inputShape: Shape): AbstractModule[Tensor[T], Tensor[T], T]

    Permalink
    Definition Classes
    Recurrent → Recurrent → KerasLayer
  21. final def eq(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  22. def equals(other: Any): Boolean

    Permalink
    Definition Classes
    Container → AbstractModule → AnyRef → Any
  23. final def evaluate(): GRU.this.type

    Permalink
    Definition Classes
    Container → AbstractModule
  24. final def evaluate(dataSet: LocalDataSet[MiniBatch[T]], vMethods: Array[_ <: ValidationMethod[T]]): Array[(ValidationResult, ValidationMethod[T])]

    Permalink
    Definition Classes
    AbstractModule
  25. final def evaluate(dataset: RDD[MiniBatch[T]], vMethods: Array[_ <: ValidationMethod[T]]): Array[(ValidationResult, ValidationMethod[T])]

    Permalink
    Definition Classes
    AbstractModule
  26. final def evaluate(dataset: RDD[Sample[T]], vMethods: Array[_ <: ValidationMethod[T]], batchSize: Option[Int]): Array[(ValidationResult, ValidationMethod[T])]

    Permalink
    Definition Classes
    AbstractModule
  27. final def evaluateImage(imageFrame: ImageFrame, vMethods: Array[_ <: ValidationMethod[T]], batchSize: Option[Int]): Array[(ValidationResult, ValidationMethod[T])]

    Permalink
    Definition Classes
    AbstractModule
  28. def finalize(): Unit

    Permalink
    Attributes
    protected[java.lang]
    Definition Classes
    AnyRef
    Annotations
    @throws( classOf[java.lang.Throwable] )
  29. def findModules(moduleType: String): ArrayBuffer[AbstractModule[_, _, T]]

    Permalink
    Definition Classes
    Container
  30. final def forward(input: Tensor[T]): Tensor[T]

    Permalink
    Definition Classes
    AbstractModule
  31. var forwardTime: Long

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  32. def freeze(names: String*): GRU.this.type

    Permalink
    Definition Classes
    Container → AbstractModule
  33. def from[T](vars: Variable[T]*)(implicit arg0: ClassTag[T], ev: TensorNumeric[T]): Variable[T]

    Permalink

    Build graph: some other modules point to current module

    Build graph: some other modules point to current module

    vars

    upstream variables

    returns

    Variable containing current module

    Definition Classes
    Net
  34. final def getClass(): Class[_]

    Permalink
    Definition Classes
    AnyRef → Any
  35. def getExtraParameter(): Array[Tensor[T]]

    Permalink
    Definition Classes
    Container → AbstractModule
  36. def getGradHiddenState(): Activity

    Permalink
    Definition Classes
    Recurrent
  37. def getHiddenShape(): Array[Int]

    Permalink
    Definition Classes
    Recurrent
  38. def getHiddenState(): Activity

    Permalink
    Definition Classes
    Recurrent
  39. final def getInputShape(): Shape

    Permalink
    Definition Classes
    InferShape
  40. final def getName(): String

    Permalink
    Definition Classes
    AbstractModule
  41. final def getNumericType(): TensorDataType

    Permalink
    Definition Classes
    AbstractModule
  42. final def getOutputShape(): Shape

    Permalink
    Definition Classes
    InferShape
  43. def getParametersTable(): Table

    Permalink
    Definition Classes
    Container → AbstractModule
  44. final def getPrintName(): String

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  45. final def getScaleB(): Double

    Permalink
    Definition Classes
    AbstractModule
  46. final def getScaleW(): Double

    Permalink
    Definition Classes
    AbstractModule
  47. def getTimes(): Array[(AbstractModule[_ <: Activity, _ <: Activity, T], Long, Long)]

    Permalink
    Definition Classes
    Container → AbstractModule
  48. final def getTimesGroupByModuleType(): Array[(String, Long, Long)]

    Permalink
    Definition Classes
    AbstractModule
  49. final def getWeightsBias(): Array[Tensor[T]]

    Permalink
    Definition Classes
    AbstractModule
  50. var goBackward: Boolean

    Permalink

    Whether the input sequence will be processed backwards.

    Whether the input sequence will be processed backwards. Default is false.

  51. val goBackwards: Boolean

    Permalink
    Definition Classes
    Recurrent → Recurrent
  52. var gradInput: Tensor[T]

    Permalink
    Definition Classes
    AbstractModule
  53. final def hasName: Boolean

    Permalink
    Definition Classes
    AbstractModule
  54. def hashCode(): Int

    Permalink
    Definition Classes
    Container → AbstractModule → AnyRef → Any
  55. val innerActivation: KerasLayer[Tensor[T], Tensor[T], T]

    Permalink

    Activation function for inner cells.

    Activation function for inner cells. You can also pass in corresponding string representations such as 'relu' or 'sigmoid', etc. for simple activations in the factory method. Default is 'hard_sigmoid'.

  56. val inputShape: Shape

    Permalink
    Definition Classes
    Recurrent → Recurrent
  57. def inputs(first: (ModuleNode[T], Int), nodesWithIndex: (ModuleNode[T], Int)*): ModuleNode[T]

    Permalink
    Definition Classes
    KerasLayer → AbstractModule
  58. def inputs(nodes: Array[ModuleNode[T]]): ModuleNode[T]

    Permalink
    Definition Classes
    KerasLayer → AbstractModule
  59. def inputs(nodes: ModuleNode[T]*): ModuleNode[T]

    Permalink
    Definition Classes
    KerasLayer → AbstractModule
  60. def isBuilt(): Boolean

    Permalink
    Definition Classes
    KerasLayer → InferShape
  61. def isFrozen[T]()(implicit arg0: ClassTag[T]): Boolean

    Permalink
    Definition Classes
    Net
  62. final def isInstanceOf[T0]: Boolean

    Permalink
    Definition Classes
    Any
  63. def isKerasStyle(): Boolean

    Permalink
    Definition Classes
    KerasLayer → InferShape
  64. final def isTraining(): Boolean

    Permalink
    Definition Classes
    AbstractModule
  65. def labor: AbstractModule[Tensor[T], Tensor[T], T]

    Permalink
    Definition Classes
    KerasLayer
  66. def labor_=(value: AbstractModule[Tensor[T], Tensor[T], T]): Unit

    Permalink
    Definition Classes
    KerasLayer
  67. var line: String

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  68. final def loadModelWeights(srcModel: Module[Float], matchAll: Boolean): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  69. final def loadWeights(weightPath: String, matchAll: Boolean): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  70. var mInputShape: Shape

    Permalink

    A Single Shape, does not include the batch dimension.

  71. val modules: ArrayBuffer[AbstractModule[Activity, Activity, T]]

    Permalink
    Definition Classes
    Container
  72. final def ne(arg0: AnyRef): Boolean

    Permalink
    Definition Classes
    AnyRef
  73. final def notify(): Unit

    Permalink
    Definition Classes
    AnyRef
  74. final def notifyAll(): Unit

    Permalink
    Definition Classes
    AnyRef
  75. var output: Tensor[T]

    Permalink
    Definition Classes
    AbstractModule
  76. val outputDim: Int

    Permalink
    Definition Classes
    Recurrent → Recurrent
  77. var outputDimension: Int

    Permalink

    Hidden unit size.

    Hidden unit size. Dimension of internal projections and final output.

  78. def parameters(): (Array[Tensor[T]], Array[Tensor[T]])

    Permalink
    Definition Classes
    Container → AbstractModule
  79. final def predict(dataset: RDD[Sample[T]], batchSize: Int, shareBuffer: Boolean): RDD[Activity]

    Permalink
    Definition Classes
    AbstractModule
  80. final def predictClass(dataset: RDD[Sample[T]], batchSize: Int): RDD[Int]

    Permalink
    Definition Classes
    AbstractModule
  81. final def predictImage(imageFrame: ImageFrame, outputLayer: String, shareBuffer: Boolean, batchPerPartition: Int, predictKey: String, featurePaddingParam: Option[PaddingParam[T]]): ImageFrame

    Permalink
    Definition Classes
    AbstractModule
  82. def processInputs(first: (ModuleNode[T], Int), nodesWithIndex: (ModuleNode[T], Int)*): ModuleNode[T]

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  83. def processInputs(nodes: Seq[ModuleNode[T]]): ModuleNode[T]

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  84. final def quantize(): Module[T]

    Permalink
    Definition Classes
    AbstractModule
  85. val rec: InternalRecurrent[T]

    Permalink
    Definition Classes
    Recurrent
  86. def release(): Unit

    Permalink
    Definition Classes
    Container → AbstractModule
  87. def reset(): Unit

    Permalink
    Definition Classes
    Container → AbstractModule
  88. def resetTimes(): Unit

    Permalink
    Definition Classes
    Container → AbstractModule
  89. var returnSeq: Boolean

    Permalink

    Whether to return the full sequence or only return the last output in the output sequence.

    Whether to return the full sequence or only return the last output in the output sequence. Default is false.

  90. val returnSequences: Boolean

    Permalink
    Definition Classes
    Recurrent → Recurrent
  91. final def saveCaffe(prototxtPath: String, modelPath: String, useV2: Boolean, overwrite: Boolean): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  92. final def saveDefinition(path: String, overWrite: Boolean): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  93. final def saveModule(path: String, weightPath: String, overWrite: Boolean): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  94. final def saveTF(inputs: Seq[(String, Seq[Int])], path: String, byteOrder: ByteOrder, dataFormat: TensorflowDataFormat): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  95. final def saveTorch(path: String, overWrite: Boolean): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  96. final def saveWeights(path: String, overWrite: Boolean): Unit

    Permalink
    Definition Classes
    AbstractModule
  97. var scaleB: Double

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  98. var scaleW: Double

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  99. final def setExtraParameter(extraParam: Array[Tensor[T]]): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  100. def setGradHiddenState(gradHiddenState: Activity): Unit

    Permalink
    Definition Classes
    Recurrent
  101. def setHiddenState(hiddenState: Activity): Unit

    Permalink
    Definition Classes
    Recurrent
  102. final def setLine(line: String): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  103. final def setName(name: String): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  104. def setScaleB(b: Double): GRU.this.type

    Permalink
    Definition Classes
    Container → AbstractModule
  105. def setScaleW(w: Double): GRU.this.type

    Permalink
    Definition Classes
    Container → AbstractModule
  106. final def setWeightsBias(newWeights: Array[Tensor[T]]): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
  107. final def synchronized[T0](arg0: ⇒ T0): T0

    Permalink
    Definition Classes
    AnyRef
  108. def toGraph(startNodes: ModuleNode[T]*): Graph[T]

    Permalink
    Definition Classes
    AbstractModule
  109. def toString(): String

    Permalink
    Definition Classes
    AbstractModule → AnyRef → Any
  110. var train: Boolean

    Permalink
    Attributes
    protected
    Definition Classes
    AbstractModule
  111. final def training(): GRU.this.type

    Permalink
    Definition Classes
    Container → AbstractModule
  112. var uRegularizer: Regularizer[T]

    Permalink

    An instance of Regularizer, applied the recurrent weights matrices.

    An instance of Regularizer, applied the recurrent weights matrices. Default is null.

  113. def unFreeze(names: String*): GRU.this.type

    Permalink
    Definition Classes
    Container → AbstractModule
  114. def updateGradInput(input: Tensor[T], gradOutput: Tensor[T]): Tensor[T]

    Permalink
    Definition Classes
    KerasLayer → AbstractModule
  115. def updateOutput(input: Tensor[T]): Tensor[T]

    Permalink
    Definition Classes
    KerasLayer → AbstractModule
  116. var wRegularizer: Regularizer[T]

    Permalink

    An instance of Regularizer, (eg.

    An instance of Regularizer, (eg. L1 or L2 regularization), applied to the input weights matrices. Default is null.

  117. final def wait(): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  118. final def wait(arg0: Long, arg1: Int): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  119. final def wait(arg0: Long): Unit

    Permalink
    Definition Classes
    AnyRef
    Annotations
    @throws( ... )
  120. def zeroGradParameters(): Unit

    Permalink
    Definition Classes
    AbstractModule

Deprecated Value Members

  1. final def save(path: String, overWrite: Boolean): GRU.this.type

    Permalink
    Definition Classes
    AbstractModule
    Annotations
    @deprecated
    Deprecated

    (Since version 0.3.0) please use recommended saveModule(path, overWrite)

Inherited from Net

Inherited from Recurrent[T]

Inherited from bigdl.nn.keras.Recurrent[T]

Inherited from KerasLayer[Tensor[T], Tensor[T], T]

Inherited from Container[Tensor[T], Tensor[T], T]

Inherited from AbstractModule[Tensor[T], Tensor[T], T]

Inherited from InferShape

Inherited from Serializable

Inherited from Serializable

Inherited from AnyRef

Inherited from Any

Ungrouped